Main
Aaron Miles
I am passionate about putting data into use. Whether by creating a model for a machine to action against, or a dashboard enabling a coworker to better understand something; my ultimate goal is to help get the most out of existing data. I have experience in all phases of the data science cycle, from scraping data from the web, to putting packaged applications into production. Hard problems are my favorite because they give me the opportunity to deeply think through issues and processes and address them in ways no one has yet thought of.
Industry Experience
Artificial Intelligence Engineer
Covail
Columbus, OH
2020 - 2018
- Lead data science and machine learning client engagements
- Developed a natural language classifier, used in multiple engagements across multiple clients
- Packaged custom solutions and deployed them in multiple ways, depending on client need. Including a hosted R package, Alteryx node, API endpoint, SQL Query, and others.
Data Scientist
The Shipyard
Columbus, OH
2018 - 2017
- Built automated ad bidding technology using machine learning to optimize ad spend
- Presented A/B test results to clients
- Developed a way to measure site template effectiveness using image processing techniques
Data Scientist
Express
Columbus, OH
2017 - 2015
- Constructed a new customer retention model, and presented on a regular cadence to C-Suite executives
- Prototyped customer segmentation, media fix, and forecasting models
- Assisted merchants with ad-hoc requests as needed
Education
M.A. Sociology
The Ohio State University
Columbus, OH
2015 - 2013
- Thesis: Evaluating the effectiveness of schools on students from different socioeconomic backgrounds
- Graduate Research Fellowship
B.S. Sociology
Brigham Young University
Columbus, OH
2013 - 2009
- Dean’s List
Research Experience
Research Assistant
Center for Human Resource Research
The Ohio State University
2015 - 2014
- Curated dataset of all Ohio public schools. Assisted development of HS Graduation model
Selected Data Science Writing
Case Study in Quanitfying Risk and Reward with Tensorflow Probability
amiles.netlify.com
N/A
2019
- Showcase the power of Tensorflow Probability in finance using Lendingclub data
Building Optimal Daily Fantasy Lineups in R
amiles.netlify.com
N/A
2019
- Used R to scrape fantasy football projections and build optimal lineups
- Used linear optimization to build lineups
Exploring Stock Relationships with Tiingo News
amiles.netlify.com
N/A
2019
- Network analysis of stock news and identifying relationships between tickers.
Selected Talks
Think Probabilistically? Model Probabilistically. Applications of Tensorflow Probability using Financial Data
Big Data Finance
N/A
2020 - 2020
- Presented how tensorflow probability could be used with financial data